When inoculated into susceptible strains of neonatal mice, the Fr98 strain of polytropic murine retrovirus induces a severe brain disease with ataxia, seizures and death in about 3 weeks. We compared Fr98 to another closely related polytropic mouse retrovirus, Fr54, which infects the same microglial and endothelia cells as Fr98, but causes no clinical disease. These viruses differ structurally only in sequences encoding the envelope gene, and generation of several recombinant chimeric viruses has shown that there are two separate regions of the viral envelope which mediate separate and cooperative mechanisms of disease induction. In spite of their differences in disease induction these viruses both induce a similar brain pathology consisting of microglial nodules, microglial giant cells, and astrogliosis with no histopathological evidence for neuronal killing or lymphocyte infiltration. Analysis of expression levels of many host gene families by RNase protection assays found upregulation of certain cytokine and chemokine genes including MCP-1, RANTES, MIP-1alpha, MIP-1beta, TNFalpha and IP-10. In contrast to typical viral models of CNS disease there was no alteration of genes involved in neuronal cell death, nitric oxide synthase or peripheral immune responses. Chemokines and cytokines might be upregulated as a host response to the damage induced or they might be directly involved in the causing of the damage. To distinguish between these alternatives chemokine receptor knockout mice lacking CCR2 or CCR5 expression were studied. Mice lacking CCR2 showed a delayed induction of clinical disease whereas mice lacking CCR5 showed no alteration in disease kinetics. Analysis of the chemokines using CCR2 as a receptor showed that MCP-1, MCP-3 and MCP-5 were upregulated in Fr98 infected mice prior to or during disease. Furthermore, direct CNS inoculation of antibodies to these chemokines indicated that anti-MCP-1 alone was able to block or delay disease induction. By in situ hybridization MCP-1 was found to be expressed by uninfected activated astrocytes. These data demonstrate an important role for uninfected astrocytes in an indirect pathogenic mechanism in this model. Since MCP-1 is also known to be upregulated in HIV- induced brain disease, which has a similar pathology to Fr98, it is possible that MCP-1 may also play a causative role in this important human CNS disease.